Inside ESID

Undertaking PSR research in Uganda

2 March 2015
By Badru Bukenya.
Research projects that rely on elite/expert interviews with government bureaucrats and politicians are quite cumbersome to execute (Lilleker, 2003; Morris, 2009), particularly in developing country contexts. In Uganda most bureaucrats would prioritise activities that directly bring in finances. Secondly, they are suspicious of researchers and fear being misquoted. It is even an uphill task for the local researchers to access their compatriots, not least due to the perceived power imbalance in favour of the bureaucrats/politicians. These factors bring to the fore the challenges of gaining access to such reluctant respondents. As our experience with Public Sector Reform (PSR) research in Uganda shows, going around these issues calls for innovative strategies, some of which may be considered illegitimate and/or slightly manipulative in the arena of scientific research.

PSR research in Uganda

I set out to roll out ESID’s PSR study in my capacity as the Uganda country researcher, after having recently been hired as an ESID Research Associate, and having undergone a thorough methodological training for this project in South Africa. However, I soon realised that the job was not going to be as straightforward as originally thought.
The initial idea was to build uptake activities into research early in the process. For Uganda, it was planned that key research stakeholders would be invited for a round table discussion prior to fieldwork. Indeed, discussions with key people in donor agencies like DFID and World Bank showed that this was a good move, for it would ensure that the research takes care of the interests of various stakeholders.
However, on the part of government, most of the officials contacted about the project were not enthusiastic about spending their valuable time on this. Therefore I decided to push ‘uptake’ to the later stages of the research and instead focus on doing actual interviews. This was not fruitful either: of the seven state agencies contacted via email, only one responded. A quick diagnosis of this lukewarm response pointed to the fact that I represented myself as an independent researcher – a position rarely respected in the country, save for established high-profile researchers.

Getting institutional affiliation

The solution to this predicament lay in getting institutional affiliation; however, at first I did not know the right organisation to partner with. I first toyed with the idea of being affiliated to a civil society organisation known for its political advocacy in Kampala. However, I quickly found that this strategy would backfire, as bureaucrats are suspicious of CSOs and consider them as belonging to the ‘opposition’. The second organisation was a research institute that had already done a couple of research projects with the University of Manchester. However, efforts to bring this one on board were rendered futile, as officials there seemed uninterested. I was told, informally, that Manchester should have contacted them directly if there was need for their collaboration on this particular research.
It was at this point that I figured out something that provided a breakthrough – working with the Political Science Department at Makerere University. The affiliation was strategic, because many of the public sector studies in Uganda have historically been done by members of staff from this department. Although I worked in a different university department, when approached the Head of Political Science gladly welcomed the idea of us working together. She quickly got one of her senior staff members interested in working on the project. It was then agreed that the three of us needed to come up with a new strategy.

A research/uptake strategy adapted to context

The first thing we agreed upon was the need to come up with a study matrix which stipulated the government agencies of interest, key research questions, and the nature of the information we needed from them.
Secondly, we needed to send introduction letters, signed by the Head of Department, to all the identified government agencies and/or officials. This was considered a useful approach, because Makerere University is a public institution and therefore was likely to be respected by other government agencies. Our letter covered four main themes:

1. What the research project is about;
2. Main institutions involved;
3. The importance of interviewing the identified office;
4. An appendix of the key questions.

We found that sending the list of questions in advance was very useful, as it “gives the impression that you are conducting a well thought out piece of research and also allows them time to consider your questions and prepare answers” (Lilleker, 2003:209).
Thirdly, we agreed, where possible, to use personal contacts with friends and past workmates within the targeted agencies as the entry point into these ministries. The new team members were better networked and had contacts with high ranking bureaucrats, which made this strategy extremely useful.
Lastly, we agreed that for every interview we secured we would use the interaction to mobilise respondents for uptake activities now scheduled for the last stage of the project. Indeed, all the respondents we talked to seemed interested in participating in feedback sessions to discuss the draft research report. The lesson here is that respondents are keen to participate in feedback events after the research process, perhaps with the view of ascertaining that their views are accurately captured in the report.

Public officials: benevolent, indifferent, avoider

In the end we managed to secure interviews in all the government units of interest. However, it is important to note that we still faced three types of bureaucrats and/or politicians: the benevolent, the indifferent, and the avoiders. The first category involved those officials who, on receiving the letter from the university, were excited to talk, sometimes revealing a lot more information than we would ask for. Indifferent officials agreed to speak, but were guarded in their responses; they often talked about how things are supposed to be, as opposed to how they actually are. Lastly, avoiders were officials who still proved elusive, despite receiving our letter; they were particularly common in one Ministry recently implicated in a major corruption scandal.
Obtaining meaningful information from these varied personalities poses a challenge that will only be overcome by our creativity as researchers.
References
Lilleker, D. G. (2003). ‘Interviewing the political elite: navigating a potential minefield’. Politics, 23(3), 207–214..
Morris, Z. S. (2009). ‘The truth about interviewing elites’. Politics, 29(3), 209–217.